DE19855250A1 - Process for optimizing an X-ray image - Google Patents

Abstract

The invention relates to a method for image optimization of an x-ray image. DOLLAR A The object of the invention to demonstrate a method for image optimization of X-ray images that automatically detects objects that are difficult to identify and minimizes the time required for this detection is achieved in that a region (A) of the X-ray image that appears dark is automatically determined and this region (A ) is brightened locally. This dark area (A) is caused by the absorption property of the object (7.1) to be illuminated. After a local image optimization, these dark areas (A) are displayed on a monitor (5) in a lightened manner, without the lighter remaining areas also being illuminated. The dark areas (A) are localized by spreading the gray value values of the pixels from which the x-ray image is composed. By brightening the areas (A), an object (7.2) underneath can now also be detected.

Description

The invention relates to a method for image optimization of an x-ray image according to the preamble of claim 1.

With the help of X-rays, objects are illuminated and the X-ray image for one Operator made visible. The objects usually contain other objects.

When the objects are illuminated, the X-rays are attenuated differently and this attenuated radiation can be seen on monitors using appropriate devices made. The X-ray images consist of pixels with different properties, such as gray value and material value.

An image processing method for material recognition is in the unpublished DE 198 12 055.9. This is done during partial screening of the luggage Detected signals as image data divided into image strips, read in and temporarily saved. Often dark areas appear in these image strips.

In a known manner, this dark area of the x-ray image is therefore identified by a Operator evaluated visually, for which purpose various optimization functions through manual Operation can be selected. This process is time-consuming and especially by hand baggage screening, such as at airports, undesirable.

A method for operating an automatic X-ray exposure device is disclosed in DE 43 30 787 A1. Here, a method is created that allows automatic selection of the measuring field enables. A first image computer calculates the gray value distribution on a test image by superimpose these gray values on a main image that is subsequently generated. This is supposed to optimal exposure within the gray value ranges. For a baggage screening is the use of a test image is too time-consuming.  

A method and a device for dynamic compression of multi-level gray-scale images is disclosed in DE 44 09 790 A1. The input image values that are below a selectable approximately medium signal threshold, inverted. The image values above as Absolute values with regard to this signal threshold are retained. The result of this Implementation of the input image values is an inlaid positive / negative image with one Monotonous jump of the gray value display clearly marked closed line and an extra halved dynamic range. This procedure is preferred in mammography applied. For use in evaluating many x-rayed materials in the luggage this procedure cannot be used.

The object of the invention is a method for image optimization of x-ray images to show the areas that are automatically difficult to identify optimally and thus the Time for identification minimized.

The object is achieved by the features of patent claim 1.

The invention is based on the idea of regions that are difficult to identify, for example as areas that appear dark, automatically determine the x-ray image and localize this area to lighten. These dark areas are illuminated using a monitor, without the lighter remaining areas are also illuminated The local lightening of the dark areas takes place by spreading the gray values of the pixels from which the x-ray image is made is composed. By lightening the areas, there are also hidden objects detectable.

Advantageous designs are contained in the subclaims.

The optimized method is only included in the image evaluation when a detected one Gray value falls below a certain threshold, or is equal to or greater than this Threshold value and if a certain number of pixels that are directly adjacent be determined. However, it is taken into account that, for example, areas that are too small are a local brightening or image optimization should not cause, because it is too confusing X-rays can come  Another option is to not directly adjacent pixels that a certain Fall below the gray value threshold to determine the number of pixels with this property to be used, but also pixels that fall below a certain gray value threshold and are not directly adjacent, for example 2 or 3 pixels can be in between, that are above the gray value threshold. Noisy areas can also be detected, if they are interspersed with some bright pixels.

Another option is to only change every gray or white pixel every third or fourth pixel to be checked, whereby it is checked whether this exceeds the gray value threshold value or not. If the gray value of the 3rd falls below the gray value threshold and is therefore the same size as that previously evaluated pixel gray value, the omitted pixels are added up 1, 2 (or 1 to 3) to create an entire image. However, there are bigger differences, the last 3 or 4 pixels are scanned again and evaluated. This will make the Processing speed in the system increased.

Since in addition to the absorption values, material values are also determined by the detectors , it is also possible to create a material-dependent image optimization. In doing so the pixels counted, scanned for a special property, thereby for gray value values searched and this scanned area lightened.

The invention will be explained in more detail using an exemplary embodiment with a drawing.

It shows:

Fig. 1 shows a simplified measuring arrangement;

Fig. 2 is a block diagram representation of the optimization procedure;

FIG. 3a shows a monitor display image without optimization;

FIG. 3b shows a monitor display with image enhancement;

Fig. 4 shows a further measuring arrangement shown in simplified form.

In Fig. 1 a measuring arrangement is illustrated with an X-radiation generator here in an irradiation device 1 and shown with a detector device 2 in simplified form. An object 3 to be determined is located between the detector device 2 and the radiation device 1 . This object 3 can be a suitcase in which various objects 7.1 , 7.2 are arranged, whereby the object 7.2 can be completely covered by the object 7.1 . A computer system 4 is connected to the detector device 2 via known components (not shown here). The measurement results are visualized via a display device, for example via a monitor 5 or a printer 6 , which are connected to the computer system 4 .

In FIG. 2, the internal construction of the computer system 4 is shown in block imagewise. The connections of the detector device (not shown here) are guided to an area determination device 8 . The outputs of the area determination device 8 are connected to a loading area optimization device 9 , the output of which is connected, for example, to the monitor 5 .

The image data is synchronized, where necessary, by means of a buffer memory (not shown). The individually listed assemblies are combined in an image optimization unit ( 10 ).

The image optimization process runs as follows.

An X-ray radiation is brought from the radiation device 1 as an X-ray beam FX 1 onto the object 3 to be illuminated. This X-ray radiation FX1 is attenuated by the respective absorption behavior of the material of the objects in the object 3 ( 7.1. , 7.2 ) and by the housing material of the object 3 and picked up by the detector device 2. The detector device 2 , for example a line camera consisting of several X-ray detectors, delivers from the non-absorbed x-rays, signals which are fed into the computer system 4 as image data information about the illuminated object 3 for image processing. This feed is preferably carried out line by line and continuously. The image data are fed into the area determination device 8 (BBE). Areas (A) are searched there, for example, by comparing the gray values of the image data against a gray value threshold. The areas can be determined by comparing the individual pixels or by combining pixels. Area sizes below a certain threshold are discarded.

At the same time, a corresponding function for optimization can be determined in the BBE ( 8 ) (for example a gray value adjustment via a histogram compensation).

The image data and, if present, the optimization function are transferred to the area optimization device (BOE) 9 . There the local image area is optimized according to the previously selected optimization function. The optimization function can either be stored in the area optimization device 9 or loaded dynamically depending on the image data of the area from the area determination device ( 8 ) into the area optimization device 9 .

In FIG. 3a, for example, object 3 is depicted with a highly absorbent object 7.1 . The image data are fed into the BBE ( 8 ) line by line. There the pixels are scanned for gray values.

In practice, the gray value range of an X-ray image is between 0 and 4095. From bright ones Areas are spoken with a gray value of 800 or higher, dark areas if they are smaller 800, with a contrast-rich distinction within the range from approx. 200 dark areas is no longer possible.

Since, as can be seen in FIG. 3, no object with strong absorption is detected in the object 3 to be illuminated until the entry of the object 7.1 , this area is classified as a bright area. The image optimization process does not need to be switched on in the system.

A gray value, which is less than 200, for example, is determined when running in at the beginning of the object 7.1 . This gray-scale value is stored, for example, as a gray-scale threshold in the area determination device 8 and now has the effect that the area of the absorbent article 7.1 is determined. For this purpose, neighboring pixels are counted and their gray value values are determined. If these gray value values are smaller than the set gray value threshold value, a dark area A is determined.

The actual image optimization is then carried out by spreading the gray value values of the individual pixels into a higher pixel value in the area optimization device 9 . This can be done using the look-up table optimization function, for example high-look-up table or other known algorithms. For example, the high look-up table can contain the following spread values:
The gray value 0 remains 0, the gray value 12 is increased or expanded into a pixel value 512, the gray value 20 into a pixel value 768, the gray value 32 into a pixel value 2944, while the pixel value 4095 remains in its value. One speaks here of a non-linear spread.

Due to the spreading of the gray value values in the selected area A, each image point is brightened in this area A, which enables high-contrast image reproduction in this image area A, which is detected as dark image area A. Such an illustration can be seen in FIG. 3b, the same area A being shown in FIG. 3a without brightening. The lightening of the dark area A now makes it possible to detect the object 7.2 , since this object 7.2 again causes darker gray values of the pixels within the same area A due to its absorption property. The areas not detected by the area determination device 8 remain in their original brightness representation when the x-ray image is displayed on the monitor 5 or on the printer 6 . There is only local brightening of the areas A detected as dark.

In addition to evaluating the properties of gray values of a pixel, image optimization can also take place on the basis of the material properties of the pixel. As shown in FIG. 4, two X-ray radiation beams FX1 and FX2, which are preferably integrated in a radiation device 1 , are brought onto the object 3 to be illuminated, for example. These X-ray radiation FX1 and FX2 have different energy ranges and are also attenuated by the respective absorption behavior of the different materials of the objects 7.1 , 7.2 and by the housing material of the object 3 and recorded by the detector device 2 . Using these two X-ray radiations FX1 and FX2 can the particular material of the individual articles 7. 1, 7.2 are detected and defined in known manner.

It is also possible to realize this with only one X-ray Materialdetektierung FX1 and a plurality of detectors, one behind the other detector means. 2 If, for example, aluminum is set in the BBE ( 8 ) as a material to be optimized, the object 7.1 is detected as area A if it consists of aluminum. The image is then optimized by spreading the gray value values as described in the area optimization device 9 .

However, several image areas A with different properties can also be determined. Different optimization functions adapted to the properties are then applied to these  applied simultaneously. The selected image areas A can additionally by frames, flashing frame or flashing between the original and optimized image area A as modified be marked.

It is understood that changes are possible within the scope of the inventive idea.

A variant of the method consists in not directly neighboring pixels, the one fall below a certain gray value threshold to determine the number of pixels with this Property, but also pixels that have a certain gray value threshold undershoot and are not directly adjacent, for example 2 or 3 pixels can lie in between, which are above the gray value threshold. It can also be used for noise Areas are detected when they are penetrated by some bright pixels.

A further variant of the method is likewise not to scan and evaluate each neighboring pixel, but rather a neighbor with m pixels. When determining this distance m, it should be noted that it must not be too large. It is important that a smaller object 7.2 located under the object 7.1 can be detected. The distance can preferably be m = 4 pixels. If, when evaluating a last scanned pixel, a gray-scale value is determined which is the same size as the previously determined one and thus also falls below the set gray-value threshold value, the pixel of the nearest distant neighbor is evaluated. However, if the report determination device 8 determines that the gray value determined is significantly greater than the last scanned pixel and thus the gray value threshold is not undershot, the pixels in the last interval are scanned again and evaluated in order to determine the exact position of the object 7.1 . After that, each pixel is either evaluated again or the interval pixel evaluation is also activated. This can increase the processing speed.

The gray value threshold set in the area determination device 8 can also be set such that the image optimization method is included in the X-ray image evaluation if a detected gray value is equal to or greater than this threshold value. Accordingly, other optimization functions are to be carried out in the area optimization device 9 .

This image optimization method is preferably used for line-by-line image acquisition. It is but it is also possible to subsequently save already stored x-ray images using this method edit and optimize in areas.

This method is not limited to the localization of dark areas A. It can even lighter areas can be spread in the dynamic. The set gray value threshold is then, for example, 1000.  

Reference list

1

Radiation device

2nd

Detector device

3rd

object

4th

Computer system

5

monitor

6

printer

7.1

object

7.2

object

8th

Area determining device

9

Area optimization facility

10th

Image optimization unit

Claims (9)

1. A method for image optimization of an X-ray image which images an object illuminated by X-rays, these X-rays being detected and processed as image data in a computer with an image memory to form an image, the image data consisting of pixels which enter the image memory, characterized in that that

• - the pixels are scanned for gray scale values,
• these sampled gray value values are compared with a gray value threshold value as the target value,
• - If the target value is not adhered to, the subsequent pixels are counted in terms of area in order to determine the value of an area (A) and the gray values of the pixels in this area (A), and the gray value values of the pixels in this area (A) in higher pixel values be spread so that the higher pixel values cause a local brightening of the x-ray image in this area (A).

2. The method according to claim 1, characterized in that the image optimization only is then carried out when a certain number of pixels in area A does not meet the setpoint. 3. The method according to claim 1 to 2, characterized in that the gray values are the same or smaller or larger than the target values. 4. The method according to claim 2 to 3, characterized in that the determination of Number of pixels is done by counting the directly adjacent pixels, whereby it is checked whether these pixels fall below a set gray value threshold or exceeds. 5. The method according to claim 2 to 3, characterized in that the determination of Number of pixels by counting individual, spaced apart from each other Pixels occur, whereby it is checked whether these pixels have a set gray value falls below or exceeds the threshold.   6. The method according to claim 2 to 3, characterized in that the determination of The number of pixels is obtained by counting individual pixels, checking that whether these pixels fall below a certain set gray value threshold ten or exceed and a maximum distance to a corresponding image do not exceed point. 7. The method according to any one of claims 1 to 6, characterized in that in advance material data provided from the pixels for determining the material to be lightened Area (A). 8. The method according to one or more of claims 1 to 7, characterized in that for optimization optimization functions for the selected or to optimize the image area (A) are fixed and stored in the image optimization unit ( 10 ). 9. The method according to one or more of the preceding claims, characterized records that several areas (A) are detected simultaneously, with different optimizations functions can be optimized and highlighted.